robust power estimation and simultaneous switching noise ... · robust power estimation and...
TRANSCRIPT
![Page 1: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/1.jpg)
Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning
March 20th, 2019
![Page 2: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/2.jpg)
2
Robust Simultaneous Switching Noise Prediction for Test using Deep Neural Network
Seyed Nima Mozaffari, Bonita Bhaskaran, Kaushik Narayanun
Ayub Abdollahian, Vinod Pagalone, Shantanu Sarangi
RTL-Level Power Estimation Using Machine Learning
Mark Ren, Yan Zhang, Ben Keller, Brucek KhailanyYuan Zhou, Zhiru Zhang
![Page 3: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/3.jpg)
3
Yuan Zhou, Zhiru Zhang
Robust Simultaneous Switching Noise Prediction for Test using Deep Neural Network
Seyed Nima Mozaffari, Bonita Bhaskaran, Kaushik Narayanun
Ayub Abdollahian, Vinod Pagalone, Shantanu Sarangi
![Page 4: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/4.jpg)
4NVIDIA CONFIDENTIAL. DO NOT DISTRIBUTE.
DFT – A BIRD’S EYE VIEW
• At-Speed Tests – verify performance
• Stuck-at Tests – detect logical faults
• Parametric Tests – verify AC/DC parameters
• Leakage Tests – catch defects that cause high leakage
Images – National Applied Research Laboratories
![Page 5: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/5.jpg)
5NVIDIA CONFIDENTIAL. DO NOT DISTRIBUTE.
SCAN TEST - SHIFT
0
1
DQ
Da ta
SI
Combinational Logic
Sc an I n (SI )
Sc an Enable (SE ) = 1
Sl ow capture c lk
Test Clk
clk clk clk
Sc an O ut (SO )
Prima ry
InputsPrima ry
O utputs
DQ
Da ta
SI
DQ
Da ta
SI
![Page 6: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/6.jpg)
6NVIDIA CONFIDENTIAL. DO NOT DISTRIBUTE.
SCAN TEST - CAPTURE
0
1
DQ
Da ta
SI
Combinational Logic
Sc an I n (SI )
Sc an E nable (SE ) = 0
Sl ow capture c lk
T est Clk
clk clk clk
Sc an O ut (SO )
Prima ry
InputsPrima ry
O utputs
DQ
Da ta
SI
DQ
Da ta
SI
![Page 7: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/7.jpg)
7
TEST WASTE FROM POWER NOISE
• Power balls overheated; Scan Freq target was lowered
• Slower frequency → Test Cost
• Higher Vmin issue
• Vmin thresholds had to be raised; impacts DPPM.
• During MBIST, overheating was observed
• Serialized tests; increase in Test Time & Test Cost
• Vmin issues observed and being debugged 0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50
60
70
80
90
100
Nominal Test
Norm
alized D
om
inan
t fc
lk
%
Norm
alized V
dd %
Voltage Freq
Linear (Voltage) Linear (Freq)
![Page 8: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/8.jpg)
8
CAPTURE NOISELow Power Capture Controller
JTAG
SCA
N IN CG-0
CP
E Q
FF FF FF
CG-1CP
E Q
FF FF FF
CG-15
CP
E Q
FF FF FF
LPC
CONTROLLER
TD_0
TD_1
TD_15
TD_2
CG-2CP
E Q
FF FF FF
![Page 9: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/9.jpg)
9
TEST NOISE ESTIMATIONThe traditional way
Power noise during test <= functional
budget directly impacts test quality !
Pre-Silicon Estimation
IR Drop
Analysis
• Can simulate only a handful of vectors
• Not easy to pick top IR-Drop inducing test patterns always
• Machine Time to simulate 3000 patterns is 6-7 years!
• Measurement is feasible for 3-5K patterns
Post-Silicon Validation
ATE Input files
Hardware & Test Program
Dev
Post-Processing
Noise per pattern
Issues
![Page 10: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/10.jpg)
10
IMPORTANCE
0
10
20
30
40
50
60
70
80
90
100
TEST C
OV
ERA
GE (%
)
TEST TIME (mS)
Test Coverage vs Test Time
LPC42
LPC73
LPC105
LPC17%
LPC40%
Strategy – we pick conservative
LPC settings!
• Higher Test Time → Higher Test Cost
• For example - Test Time savings of 40% could have been achieved.
t1 t2
LPC7%
![Page 11: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/11.jpg)
11
• Labeled data is available
• Precision is not the focus
• Need a prediction scheme that encompasses the entire production set
Why is Deep Learning a good fit?
![Page 12: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/12.jpg)
12
• Design Flow
• Feature Engineering
• Deep Learning Models
• Classification and Regression
PROPOSED APPROACH
![Page 13: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/13.jpg)
13
• Design Flow
• Feature Engineering
• Deep Learning Models
• Classification and Regression
PROPOSED APPROACH
![Page 14: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/14.jpg)
14
DESIGN FLOW
Goal:• Supervised learning model to reduce the time
and effort spent
• Most effective set of input features
Dataset:• Input features → parameters that impact the Vdroop
• Lebels → Vdroop values from silicon measurements
• Train phase → train:80% & dev:10%
• Inference phase → test:10%
Addresses the following:• Takes into account all the corner cases for PVTf
variations
• Helps predict achievable Vmin
• Cuts down post-silicon measurements – typically
6-8 weeks of engineering effort
![Page 15: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/15.jpg)
15
HARDWARE SET-UP AND SCOPESHOT
Yellow – PSN
Green – Scan Enable
Purple – CLK
Pink – Trigger
![Page 16: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/16.jpg)
16
MATLAB POST PROCESSING
• To be able to accurately tabulate the VDD_Sense droop vs. respective clock domain frequency, a Matlab script is used.
• Inputs to this script are the stored “.bin” files from the scope
• Outputs from Matlab script are:
![Page 17: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/17.jpg)
17
SNAPSHOT OF DATASET
Pattern
Global
Switch
Factor % Process Voltage Temp
Freq
(MHz) IP Name Product LPC
Droop
(mV)
Granular
Features1 2.00% 3 1 10 1000 1 2 3 30
2 3.00% 3 1 10 1000 1 2 3 35
3 3.00% 3 1 10 1000 1 2 3 35
4 4.00% 3 1 10 1000 1 2 3 35
5 3.00% 3 1 10 1000 1 2 3 33
6 2.00% 3 1 10 1000 1 2 3 33
7 60.00% 3 1 10 1000 1 2 3 100
8 45.00% 3 1 10 1000 1 2 3 85
9 65.00% 3 1 10 1000 1 2 3 105
10 36.10% 3 1 10 1000 1 2 3 60
11 36.00% 3 1 10 1000 1 2 3 61
12 33.00% 3 1 10 1000 1 2 3 60
13 50.00% 3 1 10 1000 1 2 3 90
. . .
. . .
. . .
2998 29.87% 3 1 10 1000 1 2 3 55
2999 47.84% 3 1 10 1000 1 2 3 85
3000 58.92% 3 1 10 1000 1 2 3 91
![Page 18: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/18.jpg)
18
DEPLOYMENT
Goal• Optimize low power DFT architecture
• Generate reliable test patterns
PSN analysis is repeated• at various milestones of the chip design cycle
and finalized close to tape-out.
• until there are no violations for any of the test
patterns.
![Page 19: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/19.jpg)
19
• Design Flow
• Feature Engineering
• Deep Learning Models
• Classification and Regression
PROPOSED APPROACH
![Page 20: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/20.jpg)
20
FEATURE ENGINEERING
IP-level (Global)• GSF
• PVT
• PLL frequency f
• LP_Value
• Type
SoC sub-block-level (Local)• LSF
• Instance_Count
• Sense_Distance
• Area
![Page 21: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/21.jpg)
21
EXAMPLE: FEATURE EXTRACTION
➢ on-chip measurement point location➢ sense point neighborhood-level graph➢ global and local feature vectors
Sub-Block-Level layout of an SoC
![Page 22: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/22.jpg)
22
• Design Flow
• Feature Engineering
• Deep Learning Models
• Classification and Regression
PROPOSED APPROACH
![Page 23: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/23.jpg)
23
DEEP LEARNING MODELS
Fully Connected (FC) model• basic type of neural network and is used in most of the models.
• Flattened FC model
• Hybrid FC model
Natural Language Processing-based (NLP) model• NLP is traditionally used to analyze human language data.
• we apply the concept of the averaging layer to our IR drop prediction problem.
• Model is independent of the number of sub-blocks in a chip.
![Page 24: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/24.jpg)
24
FLATTENED FC MODEL
All the input features are applied simultaneously to the first layer.
![Page 25: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/25.jpg)
25
HYBRID FC MODEL
Input features are divided into different groups, each applied to a different layer.
![Page 26: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/26.jpg)
26
NLP MODEL
➢ Local features of each sub-block form an individual bag of numbers.
➢ Filtered Average (FA): 1) filters out non-toggled sub-blocks, 2) calculates the average.
![Page 27: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/27.jpg)
27
• Design Flow
• Feature Engineering
• Deep Learning Models
• Classification and Regression
PROPOSED APPROACH
![Page 28: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/28.jpg)
28
CLASSIFICATION AND REGRESSION
➢ Classification models predict a discrete value (or a bin).
➢ Regression models predict the absolute value.
➢ Optimization:
➢ Cost Function:
➢ Loss Function: 𝐿 𝑦𝑖 , ො𝑦𝑖
Input Normalization, Adam optimizer, learning rate decay, L2 regularization
𝐽 =1
𝑚
𝑖=1
𝑚
𝐿 𝑦𝑖 , ො𝑦𝑖 + ∅(𝑤)
−(𝑦𝑖 log ො𝑦𝑖 + (1 − 𝑦𝑖) log(1 − ො𝑦𝑖)) 𝑠𝑞𝑟𝑡(1
𝑘
𝑖=1
𝑘
𝑦𝑖 − ො𝑦𝑖2)
classification regression
![Page 29: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/29.jpg)
29
RESULTS
Benchmark Information - 16nm GPU chips: Volta-IP1 and Xavier-IP2
➢ Local features are wrapped with zero-padding (only for FC)
➢ Approximately 90% of the samples for training and validation
➢ Approximately 10% of the samples for inference.
Models were developed in Python using TensorFlow and NumPy libraries.
Models were run on a cloud-based system with 2 CPUs, 2 GPUs and 32GB memory.
GPU No. of Features No. of Train Samples No. Inference Samples
Volta-IP1 323 16500 1500
Xavier-IP2 239 2500 500
![Page 30: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/30.jpg)
30
RESULTS
Dataset Model-ArchitectureTrain
Accuracy (%)
Inference
Accuracy (%)
Train Time
(minutes)
MAE
(mV)
Volta-IP1
+
Xavier-IP2
Classification-Flattened FC 94.5 94.5 10 7.30
Classification-Hybrid FC 96.0 96.0 3 6.90
Classification-NLP 92.6 92.6 80 7.46
Regression-Flattened FC 98.0 93.0 9 7.79
Regression-Hybrid FC 98.0 96.0 3 7.25
Regression-NLP 95.0 95.0 90 7.28
Method Run-Time
Pre-Silicon Simulation 416 days
Post-Silicon Validation 84 mins
Proposed 0.33 secs
Average run-time or prediction time
➢ For a 500-pattern set
![Page 31: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/31.jpg)
31
RESULTS
Correlation between the predicted and the silicon-measured Vdroop
Classification
Regression
![Page 32: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/32.jpg)
32
FUTURE WORK
• Train and apply DL for in-field test vectors noise estimation
• Shift Noise prediction
• Additional physical parameters
• Other architectures
![Page 33: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/33.jpg)
Yuan Zhou, Zhiru Zhang
RTL-Level Power Estimation Using Machine Learning
Mark Ren, Yan Zhang, Ben Keller, Brucek Khailany
![Page 34: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/34.jpg)
34
C++ SystemC
RTLGate-level
Netlist
Behavioral Level
Very fast: > 10k cycles/s(Source: [Ahuja ISQED’09] [Shao ISCA’14])
Only average power
Not that accurate
RTL Level
Slower: 1k-10k cycles/s (Source: [Yang ASP-DAC’15][PowerArtist])
Not-so-great accuracy
Some still only model
average power
Gate Level
Slowest: 10-100 cycles/s(Source: [VCS,Primetime PTPX])
Cycle-level power trace
Very accurate
Long turn-around time!
[Ahuja ISQED’09] S. Ahuja, D. A. Mathaikutty, G. Singh, J. Stetzer, S. K. Shukla, and A. Dingankar. "Power estimation methodology for a high-level synthesis framework." In Quality of Electronic Design, 2009. ISQED 2009. Quality Electronic Design, pp. 541-546. IEEE, 2009.[Shao ISCA’14] Y. Shao, B. Reagen, G.-Y. Wei, and D. Brooks. "Aladdin: A pre-RTL, power-performance accelerator simulator enabling large design space exploration of customized architectures." In 2014 ACM/IEEE 41st International Symposium on Computer Architecture (ISCA).[Yang ASP-DAC’15] J. Yang, L. Ma, K. Zhao, Y. Cai, and T.-F. Ngai. "Early stage real-time SoC power estimation using RTL instrumentation." In Design Automation Conference (ASP-DAC), 2015 20th Asia and South Pacific, pp. 779-784. IEEE, 2015.[PowerArtist] https://www.ansys.com/products/semiconductors/ansys-powerartist[VCS] https://www.synopsys.com/verification/simulation/vcs.html[Primetime PTPX] https://news.synopsys.com/index.php?item=123041
MOTIVATIONPower modeling is either slow or inaccurate.
Get power with accurate power estimation using simulation traces at early design stages?
![Page 35: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/35.jpg)
Emerging field using Machine Learning for Electronic Design Automation (EDA) tasks
Utilize GPU proficiency in ML tasks + find a way to map EDA applications to fit ML
→ Use machine learning / deep learning techniques to accurately estimate power at higher design abstraction level (RTL)
Shorter turn-around time, faster power validation, covers a diverse range of different workloads
35
OPPORTUNITY: ML FOR EDA
Source: https://towardsdatascience.com/
Source: https://roboticsandautomationnews.com/
![Page 36: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/36.jpg)
36
Simulation
Simulation Results
Power Analysis
Power Results
Gather Training Data
Feature Construction
ML Model Training
Trained Power Model
Feature EngineeringModel Training
New Test Cases
Simulation
New Simulation Results
ML Model Inference
New Power Results
Model Application
Simulation Results
Power Results
Feature Construction
TrainedPower Model
Once Once “Free”
PROPOSED SOLUTION: ML-BASED POWER ESTIMATION WORKFLOW
![Page 37: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/37.jpg)
37𝑃 = 𝐶𝑉2𝑓
Learns the amount of capacitance charging
associated with 2 1→0 transitions is possibly P
POWER ESTIMATION: CIRCUIT PERSPECTIVEOur models are essentially learning the switching capacitance associated with certain register switching activitiesFiguring out which caps switch and by how much is inhumanely complex and non linear
→ Perfect for machine learning!
Example:
![Page 38: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/38.jpg)
Traditional ML: linear model, XGBoost
With principal component analysis (PCA) applied for overfitting avoidance
Pros: smaller model, faster training
Cons: Hard to capture non-linearities
DL: convolutional neural net (CNN), multi-layer perceptron (MLP)
Pros: good for all sorts of non-linear models, good scalability
Cons: large model, longer training times, scalable but at a large startup cost (lots of parameters/nodes)
38
MODEL SELECTION
𝑃 = 𝑎0 + 𝑎1𝑥1+ 𝑎2𝑥2+ 𝑎3𝑥3+ ⋯𝑎𝑛𝑥𝑛
P1P2…Pm
a1a2…
an
= .
x1x2…
xm
Linear regression model CNN
Source:https://brilliant.org/wiki/convolutional-neural-network/
![Page 39: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/39.jpg)
What information to use?Register 0/1 state as inputs into model
How to encode? CNNs work best when features have spatial relationship for their inputsDefault (naïve) encoding: random placement of register traces in CNN inputGraph-partition based: treat register relations as a graph, then partition to determine input placementNode-embedding based: Use node2vec to convert graph nodes into embeddings (Source: [Grover
SIGKDD’16])
39
FEATURE CONSTRUCTION
[Grover SIGKDD’16] Grover, Aditya, and Jure Leskovec. "node2vec: Scalable feature learning for networks." In Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining , pp. 855-864. ACM, 2016.
a
b c
d
I
O
x
x
x
xxx
d c
b a
O I
xxx
IOad
b c
Default encoding
Graph-partition
Node-embedding
![Page 40: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/40.jpg)
Test Designs
40
EXPERIMENT SETUP
Normalized Root Mean Square Error (NRMSE)
𝑁𝑅𝑀𝑆𝐸 = 𝑅𝑀𝑆𝐸/ത𝑦
Cycle-by-cycle basis
Directly look at the power traces to see how good it fits
Good for catching outliers
Cycle-by-cycle basis
Source: Y. Zhou, et. al “PRIMAL: Power Inference using Machine Learning”, to appear in DAC 2019, June
![Page 41: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/41.jpg)
41
EXPERIMENT SETUP
ML training and inference infrastructure:
NVIDIA 1080Ti GPU
Software packages: network, metis, node2vec, Python 3.5, Keras 2.1.6, scikit-learn, xgboost 0.72.1
Ground truth and comparison baseline gate level power analysis infrastructure
Intel Xeon CPU server, 64GB RAM
![Page 42: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/42.jpg)
42
RESULTS
Good accuracy
<5% average power estimation for all test cases
CNNs outperform linear models for bigger designs
Accuracy outperforms commercial tool
Source: Y. Zhou, et. al “PRIMAL: Power Inference using Machine Learning”, to appear in DAC 2019, June
![Page 43: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/43.jpg)
~50X speedup against gate simulation + power analysis
Cycle-by-cycle traces show better accuracy for CNNs compared to linear models 43
300 cycles of RISCV core dhrystone benchmark
RESULTS
Source: Y. Zhou, et. al “PRIMAL: Power Inference using Machine Learning”, to appear in DAC 2019, June
Source: Y. Zhou, et. al “PRIMAL: Power Inference using Machine Learning”, to appear in DAC 2019, June
![Page 44: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/44.jpg)
44
CONCLUSIONS
We can get both good accuracy and high speedup with ML-based power estimation
Achieves ~50X speedup over baseline with <5% error
A good example of using ML for EDA purposes
GPUs greatly benefit training/inference time in ML for EDA
![Page 45: Robust Power Estimation and Simultaneous Switching Noise ... · Robust Power Estimation and Simultaneous Switching Noise Prediction Methods Using Machine Learning March 20th,](https://reader030.vdocuments.net/reader030/viewer/2022040204/5ed07e4548a2f70be61d34fb/html5/thumbnails/45.jpg)
Thank You!